Judy A. Mikovits, et al.: Sensitive Methods for Detecting XMRV
Distribution of Xenotropic Murine Leukemia
Virus-Related Virus (XMRV) Infection in Chronic
Fatigue Syndrome and Prostate Cancer
Judy A. Mikovits1, Ying Huang2, Max A. Pfost1, Vincent C. Lombardi1, Daniel C. Bertolette2, Kathryn S. Hagen1
and Francis W. Ruscetti2
1Whittemore-Peterson Institute for Neuroimmune Diseases, University of Nevada, Reno NV, USA; 2Laboratory of Experimental Immunology,
Cancer and Inflammation Program, National Cancer Institute-Frederick, Frederick MD, USA
AIDS Rev. 2010;12:149-52
Judy A. Mikovits
Whittemore Peterson Institute
Applied Research Facility Rm 401/MS199
1664 North Virginia St.
University of Nevada
Reno NV 89557, USA
In 2006, sequences described as xenotropic murine leukemia virus-related virus (XMRV) were
discovered in prostate cancer patients. In October 2009, we published the first direct isolation of
infectious XMRV from humans and the detection of infectious XMRV in patients with chronic fatigue
syndrome. In that study, a combination of classic retroviral methods were used including: DNA
polymerase chain reaction and reverse transcriptase polymerase chain reaction for gag and env, full
length genomic sequencing, immunoblotting for viral protein expression in activated peripheral blood
mononuclear cells, passage of infectious virus in both plasma and peripheral blood mononuclear cells
to indicator cell lines, and detection of antibodies to XMRV in plasma. A combination of these methods
has since allowed us to confirm infection by XMRV in 85% of the 101 patients that were originally
studied. Since 2009, seven studies, predominantly using DNA polymerase chain reaction of blood
products or tumor tissue, have reported failures to detect XMRV infection in patients with either
prostate cancer or chronic fatigue syndrome. A review of the current literature on XMRV supports the
importance of applying multiple independent techniques in order to determine the presence of this
virus. Detection methods based upon the biological and molecular amplification of XMRV, which is
usually present at low levels in unstimulated blood cells and plasma, are more sensitive than assays
for the virus by DNA polymerase chain reaction of unstimulated peripheral blood mononuclear cells.
When we examined patient blood samples that had originally tested negative by DNA polymerase chain
reaction by more sensitive methods, we observed that they were infected with XMRV; thus, the DNA
polymerase chain reaction tests provided false negative results. Therefore, we conclude that molecular
analyses using DNA from unstimulated peripheral blood mononuclear cells or from whole blood
are not yet sufficient as stand-alone assays for the identification of XMRV-infected individuals.
Complementary methods are reviewed, that if rigorously followed, will likely show a more accurate
snapshot of the actual distribution of XMRV infection in humans. (AIDS Rev. 2010;12:149-52)
Corresponding author: Judy A. Mikovits, firstname.lastname@example.org
Biological amplification. Molecular enrichment. XMRV. Chronic fatigue syndrome. Prostate cancer.
AIDS Reviews. 2010;12
Recently, through ViroChip analysis, sequences
corresponding to a previously unknown retrovirus were
detected in samples of human prostate tissue1. Retro-
viral DNA was isolated from a minority of the samples
in the nonmalignant (stromal) cell microenvironment.
The DNA sequences were found to be distantly related
to HIV-1 and human T-lymphotropic virus type 1 (HTLV-1),
but very closely related to a type of retrovirus named
xenotropic murine leukemia virus (X-MLV)2,3. As a result,
the newly detected virus was named xenotropic murine
leukemia virus-related virus (XMRV). Despite a high
degree of sequence identity with X-MLV, XMRV is
clearly distinct from all previously described X-MLV
and is not present as an endogenous retrovirus in
the mouse genome. Different isolates of XMRV from
prostate cancer and chronic fatigue syndrome (CFS)
patients published to date show very little sequence
variation and form a distinct branch following phylo-
genetic analysis4. Additional evidence that XMRV is a
human virus that can infect humans includes the mapping
of viral integration sites within human chromosomes5,
the presence of viral antibodies in human plasma4,6,
and the presence of viral proteins and nucleic acids in
fresh or frozen tissue4,6,7.
XMRV in prostate cancer:
positive and negative studies
The association between XMRV and prostate cancer
was strengthened in two recent reports of the presence
of XMRV proteins, as well as nucleic acid sequences,
in prostate cancer tissue7. The study by Schlaberg, et
al.7, in contrast to the original report of XMRV in stromal
cells of prostate tumors1, reported that expression of
XMRV was observed in the malignant cells in these
samples, with evidence of XMRV infection found in almost
25% of samples analyzed. The principal methods for
viral detection were either by polyclonal antisera to
XMRV virions (23% positive) or by DNA PCR (6% posi-
tive) of the integrase region. Detection of XMRV proviral
DNA via polymerase chain reaction (PCR) in this study
supported the protein data, but was less sensitive. Sub-
sequently, Arnold, et al.6 described a novel serologic
assay detecting neutralizing antibodies to XMRV, resulting
in 27.5% seropositivity in prostate cancer patients. The
authors confirmed the serologic results by PCR as well
as fluorescence in situ hybridization (FISH), another
unambiguous XMRV detection method, thus demonstrat-
ing the value of multiple independent methods.
DNA PCR has been used as the sole or primary
assay method for XMRV in several studies that failed
to detect XMRV in prostate cancer patients or tumors.
In a study from Johns Hopkins, prostate tissue DNA
from 338 prostate cancer patients was negative upon
nested PCR for gag sequences8. Two independent
studies of prostate cancer tissue samples from Germany
found no XMRV using DNA PCR, reverse transcriptase
(RT) PCR and an ELISA for XMRV Gag and Env anti-
bodies9,10. Given the discordance between such findings,
determining a role for XMRV in prostatic disease will
require more extensive epidemiological examination of
incidence in both specialized and general human
populations as well as more sensitive assays for XMRV
detection. Isolation of infectious XMRV from prostate
cancer patients has not been published.
XMRV in chronic fatigue syndrome:
positive and negative studies
Our recent study4 found that sequences to XMRV
gag and/or env were present in a high percentage
(67%) of individuals with CFS, according to either single-
round DNA PCR and/or nested RT PCR. Evidence of
XMRV infection in vivo was detected in phytohemagglu-
tinin and interleukin 2 (PHA/IL-2) activated leukocytes
isolated from peripheral blood. Further studies revealed
that patients’ T-cells grown in IL-2 and B-cells grown
in CD40L were infected in vivo with XMRV and that
T-cell lines could be infected in vitro with XMRV. More
importantly, this study also demonstrated that infectious
virus was present in infected individuals; XMRV could
be transmitted either cell-associated or cell-free from
both activated lymphocytes and plasma from infected
individuals by passage to lymph node carcinoma of
the prostate (LNCaP), a prostate cancer cell line robust
for XMRV replication11,12. Plasma from these individuals
also had antibodies specific for the envelope protein
of this type of retrovirus, showing that XMRV can elicit
an immune response. In contrast to subsequent reports,
our 2009 paper was not a survey of CFS using PCR or
a search for antibodies to XMRV proteins in patients;
instead, several different assays for the presence of
XMRV nucleic acids, proteins, and infectious virus
were performed. A list detailing the assays that were
performed on each of the samples in the initial study
has recently been published13.
Three studies from Europe published in 2010 have not
found XMRV in CFS patients. In the first study, scientists
from the United Kingdom found no gag sequences using
PCR on whole blood DNA from 186 CFS patients14.
Judy A. Mikovits, et al.: Sensitive Methods for Detecting XMRV
In the second study, PCR assays for gag and env using
peripheral blood mononuclear cell (PBMC) DNA from
UK CFS patients found all 170 samples negative by
PCR15. Groom, et al.15 also used a neutralizing viral assay
and found that 26 of the sera had neutralizing activity.
The 26 positive samples also neutralized Env proteins
of other viruses; therefore this activity is not specific. In
a Dutch study, PCR for int and gag sequences performed
on PBMC DNA of 32 CFS patients was negative16.
Examination of samples by DNA PCR gag and pol
sequences for XMRV was also negative in a recent
study reported by the Centers for Disease Control17.
Plasma from the patients and controls was also examined
for the presence of antisera that could react with MLV
proteins, with largely negative results17.
Differences between the studies
of XMRV in human populations
As detailed above, we implemented four different ways
of establishing XMRV infection in the blood products of
CFS patients. We described detection of XMRV by
single-round DNA PCR of unstimulated PBMC from our
most viremic patients4. Although it is the least sensitive
assay for XMRV, it is also the least likely to give false
positive results resulting from contamination of samples
with small amounts of mouse DNA, which contains
endogenous retroviral sequences that are similar to
XMRV. The fact that some patients exhibited readily
detectable provirus with only single-round PCR provided
strong evidence for XMRV infection.
Nested PCR is a more sensitive method than single-
round PCR for detection of low amounts of viral sequence.
Nevertheless, 96% of 210 normal samples tested in our
2009 study4 were negative upon nested PCR. Of the
101 CFS samples, 11 were positive upon single-round
DNA PCR for gag sequences. Of the 90 samples from
which no viral sequences were obtained by single-
round DNA PCR, 60 were positive for gag sequences
by nested PCR from cDNA.
There could be multiple reasons for the disparity in
results between different studies of the prevalence of
XMRV in human populations. Geographical differences
in the distribution of XMRV could account for at least
some of the discordant results. For example, the
prevalence of another retrovirus, HTLV-1, profoundly
differs around the world18,19.
Secondly, the existence of divergent XMRV or related
viruses is possible. Variant viruses could easily be
missed by many of the assays; PCR data is particularly
susceptible to sequence variation. Although a PCR assay
may work well when a known viral sequence is used as
positive control, the sensitivity of the assay may be far
less in clinical samples, where sequence variation may
confound results. Unfortunately, the primers and PCR
assays used in the reports that failed to find XMRV in
clinical samples did not utilize certified positive clinical
samples as positive controls, even though authors in
three of four negative studies were offered samples from
Lombardi, et al.4. Notably, the genetic variation between
full-length XMRV sequences currently available is 0.03%,
despite the fact that they are obtained from samples
from patients exhibiting two vastly different diseases
from geographically distinct areas. This variation is
smaller than the variation observed between HTLV-1
isolates20. Analysis of many more samples will be needed
to determine the actual extent of sequence variation in
XMRV that infects human populations.
Thirdly, whether individuals infected with XMRV form
antibodies to the viral proteins, and if so, to which viral
proteins, remains an open question. Lombardi, et al.
detected antibodies specific to XMRV Env4,13 in 20-80%
of CFS patient plasma tested vs. healthy controls in
whom the presence of antibodies to Env are between
2-7%. Further experiments need to be performed to
determine at what frequency individuals infected with
XMRV exhibit detectable antibodies to XMRV proteins;
thus, the presence or absence of antibodies to MLV
proteins cannot presently be used to determine whether
or not an individual is infected with XMRV.
Fourth, the clinical criteria for patient selection vary
widely in the CFS field. Lack of uniformity between patient
populations likely plays a major role in discordant results.
In one study, Van Kuppeweld, et al.16 employed the older
Oxford criteria21. The Oxford definition does not define
myalgic encephalomyelitis/chronic fatigue immune defi-
ciency syndrome, but instead defines idiopathic chronic
fatigue, i.e. tiredness. The patients selected in two negative
studies were diagnosed according to the 1994 Interna-
tional “Fukuda” criteria22. The criteria for the 1994 case
definition are based primarily on symptoms and not on
physical signs or chemical or immunological tests. Sug-
gestions have been made to subgroup CFS based on
clinical and immunological data23 such as those discussed
in Lombardi ,et al., which also used the most rigorous case
definition established, the Canadian Consensus Criteria
(CCC)22-24. Patients who satisfy the CCC also meet the less
rigorous Fukuda criteria, but given the heterogeneity and
complexity of the disease, are unlikely to define homoge-
nous subgroups and thus there are likely to be significant
differences within patient populations in the negative
studies, which clearly confound the interpretation of these
AIDS Reviews. 2010;12
data. To further confound the ability to draw meaningful
conclusions regarding XMRV in CFS patients, the subjects
defined as patients in Switzer, et al.17 were identified
through Georgia and Wichita registries of population-
based studies that were described as meeting criteria
clearly not that of Fukuda, et al.22. Moreover, the authors
selected study subjects that were not diagnosed by a
physician and did not satisfy the more rigorous CCC.
Additional patient data at the time of sample collection,
such as immune status, may be useful. For example,
XMRV sequences were found in the respiratory tract
secretions of 2-3% of healthy donors and 10% of immuno-
compromised donors25, suggesting that the immune status
of individuals could play a role in XMRV detection.
Despite the association of XMRV with both prostate
cancer and CFS, many questions remain regarding the
prevalence of XMRV in the human population and
the incidence of XMRV in disease. We have found that the
most sensitive blood-based assays for XMRV detection in
decreasing order are: (i) plasma or activated PBMC
co-cultured with LNCaP followed by virus isolation and
sequencing or nested PCR for gag; (ii) detection of viral
proteins in activated PBMC by flow cytometry (FACS)
or in tissue by immunohistochemistry4; (iii) or plasma
antibodies for Env; (iv) in situ hybridization by FISH6, RT
PCR in plasma or cDNA from PBMC, with direct DNA from
inactivated PBMC being the least sensitive method. We
remain confident that applying multiple methods and
rigorously following established protocols that have suc-
cessfully detected XMRV will reveal a wider distribution
of XMRV infection in humans than has currently been
reported. The study of XMRV is in its infancy and much
more information is needed concerning replication and
pathogenesis of this virus in humans. Priorities for research
are development of sensitive nucleic acid and serologic
tests for high-throughput screening and the development
of therapeutics for clinical testing. The development of
standardized, highly sensitive assays for XMRV detection
and the existence of a panel of positive clinical samples
for use as controls are clearly essential to make progress
in determining the prevalence of XMRV in various human
populations and its possible role in disease.
We would like to thank Maureen Hanson for careful
manuscript review and editing along with many helpful
1. Urisman A, Molinaro R, Fischer N, et al. Identification of a novel Gam-
maretrovirus in prostate tumors of patients homozygous for R462Q RNA-
SEL variant. PLoS Pathog. 2006;2:e25.
2. Levy J. Xenotropic viruses: murine leukemia viruses associated with NIH
Swiss, NZB, and other mouse strains. Science. 1973;182:1151-3.
3. Yan Y, Liu Q, Kozak C. Six host range variants of the xenotropic/poly-
tropic gammaretroviruses define determinants for entry in the XPR1 cell
surface receptor. Retrovirology. 2009;6:87.
4. Lombardi V, Ruscetti F, Das Gupta J, et al. Detection of an infectious
retrovirus, XMRV, in blood cells of patients with chronic fatigue syn-
drome. Science. 2009;326:585-9.
5. Kim S, Kim N, Dong B, et al. Integration site preference of xenotropic
murine leukemia virus-related virus, a new human retrovirus associated
with prostate cancer. J Virol. 2008;82:9964-77.
6. Arnold R, Makarova N, Osunkoya A, et al. XMRV infection in patients
with prostate cancer: novel serologic assay and correlation with PCR
and FISH. Urology. 2010;75:755-61.
7. Schlaberg R, Choe D, Brown K, Thaker H, Singh I. XMRV is present
in malignant prostatic epithelium and is associated with prostate
cancer, especially high-grade tumors. Proc Natl Acad Sci USA. 2009;
8. Sfanos K, Sauvageot J, Fedor H, Dick J, De Marzo A, Isaacs W. A
molecular analysis of prokaryotic and viral DNA sequences in prostate
tissue from patients with prostate cancer indicates the presence of
multiple and diverse microorganisms. Prostate. 2008;68:306-20.
9. Fischer N, Hellwinkel O, Schulz C, et al. Prevalence of human gam-
maretrovirus XMRV in sporadic prostate cancer. J Clin Virol. 2008;
10. Hohn O, Krause H, Barbarotto P, et al. Lack of evidence for xenotropic
murine leukemia virus-related virus (XMRV) in German prostate cancer
patients. Retrovirology. 2009;6:92.
11. Dong B, Silverman R. Androgen stimulates transcription and replica-
tion of xenotropic murine leukemia virus-related virus. J Virol. 2010;
12. Rodriguez J, Goff S. Xenotropic murine leukemia virus-related virus
establishes an efficient spreading infection and exhibits enhanced
transcriptional activity in prostate carcinoma cells. J Virol. 2010;
13. Mikovits J, Lombardi V, Pfost M, Hagen K, Ruscetti F. Addenda to:
Detection of an infections retrovirus, XMRV, in blood cells of patients
with chronic fatigue syndrome. Virulence. 2010;1(5):1-5.
14. Erlwein O, Kaye S, McClure M, et al. Failure to detect the novel retrovi-
rus XMRV in chronic fatigue syndrome. PLoS One. 2010;5:e8519.
15. Groom H, Yap M, Galao R, Neil S, Bishop K. Susceptibility of xenotropic
murine leukemia virus-related virus (XMRV) to retroviral restriction fac-
tors. Proc Natl Acad Sci USA. 2010;107:5166-71.
16. van Kuppeveld F, de Jong A, Lanke K, et al. Prevalence of xenotropic
murine leukaemia virus-related virus in patients with chronic fatigue
syndrome in the Netherlands: retrospective analysis of samples from an
established cohort. BMJ. 2010;340:c1018.
17. Switzer W, Jia H, Hohn O, et al. Absence of evidence of xenotropic
murine leukemia virus-related virus infection in persons with chronic
fatigue syndrome and healthy controls in the United States. Retrovirol-
18. Alcantara L, de Oliveira T, Gordon M, et al. Tracing the origin of Brazil-
ian HTLV-1 as determined by analysis of host and viral genes. AIDS.
19. Hlela C, Shepperd S, Khumalo N, Taylor G. The prevalence of human
T-cell lymphotropic virus type 1 in the general population is unknown.
AIDS Rev. 2009;11:205-14.
20. Van Dooren S, Pybus O, Salemi M, et al. The low evolutionary rate of
human T-cell lymphotropic virus type-1 confirmed by analysis of vertical
transmission chains. Mol Biol Evol. 2004;21:603-11.
21. Sharpe M, Archard L, Banatvala J, et al. A report--chronic fatigue syn-
drome: guidelines for research. J R Soc Med. 1991;84:118-21.
22. Fukuda K, Straus S, Hickie I, Sharpe M, Dobbins J, Komaroff A. The
chronic fatigue syndrome: a comprehensive approach to its definition
and study. International Chronic Fatigue Syndrome Study Group. Ann
Intern Med. 1994;121:953-9.
23. Jason L, Torres-Harding S, Jurgens A, Helgerson J. Comparing the
Fukuda et al. criteria and the Canadian definition for chronic fatigue
syndrome. J Chron Fatigue Syndr. 2004;12: 37-52.
24. Tan E, Sugiura K, Gupta S. The case definition of chronic fatigue syn-
drome. J Clin Immunol. 2002;22:8-12.
25. Fischer N, Schulz C, Stieler K, et al. Xenotropic murine leukemia virus-
related gammaretrovirus in respiratory tract. Emerg Infect Dis. 2009;